Emotion Filtering at the Edge

TL;DR

This paper presents a privacy-preserving method for emotion filtering in voice inputs at the edge using CycleGAN, effectively reducing emotional state identification while maintaining speech recognition accuracy on low-cost devices.

Contribution

The authors introduce an edge-based emotion filtering approach using CycleGAN to protect user privacy without sacrificing speech recognition performance.

Findings

Emotion identification reduced by ~91%

Speech recognition accuracy differs only ~0.16% from cloud-based methods

Effective implementation on Raspberry Pi 4

Abstract

Voice controlled devices and services have become very popular in the consumer IoT. Cloud-based speech analysis services extract information from voice inputs using speech recognition techniques. Services providers can thus build very accurate profiles of users' demographic categories, personal preferences, emotional states, etc., and may therefore significantly compromise their privacy. To address this problem, we have developed a privacy-preserving intermediate layer between users and cloud services to sanitize voice input directly at edge devices. We use CycleGAN-based speech conversion to remove sensitive information from raw voice input signals before regenerating neutralized signals for forwarding. We implement and evaluate our emotion filtering approach using a relatively cheap Raspberry Pi 4, and show that performance accuracy is not compromised at the edge. In fact, signals generated at the edge differ only slightly (~0.16%) from cloud-based approaches for speech recognition. Experimental evaluation of generated signals show that identification of the emotional state of a speaker can be reduced by ~91%.